11. Tribological Testing, Instrumentation and Failure Analysis

Introduction and Context

Tribology is ultimately validated not only by theories of friction, wear, lubrication, and surface engineering, but by the quality of the evidence used to support engineering decisions. A tribological test may be performed to compare materials, qualify a lubricant, reproduce a field failure, establish a design limit, or monitor the health of an operating machine. These purposes are related, but they do not require identical test methods. A carefully controlled laboratory test can produce precise data and still be irrelevant to service if its contact geometry, load history, temperature, lubrication regime, or surface condition differs substantially from the application.

Tribological testing therefore involves more than selecting a machine and recording a coefficient of friction. It requires definition of the tribosystem, identification of the governing failure mechanisms, control of experimental variables, calibration of instrumentation, statistical treatment of results, and interpretation of wear or surface damage in relation to service conditions. Failure analysis extends this process into the field: the investigator must distinguish symptoms from causes, preserve evidence, reconstruct operating history, and correlate observations with mechanics, materials science, lubrication, and manufacturing practice.

This chapter presents a framework for designing valid tribological experiments and conducting technically defensible failure investigations. It emphasizes the relationship between test purpose and test configuration, measurement uncertainty and reproducibility, surface and lubricant characterization, and the use of digital tools in modern condition assessment.

Learning Objectives

After studying this chapter, the reader should be able to:

1. Define the tribosystem and translate service conditions into laboratory test variables.

2. Select suitable tribometers, sensors, and relevant standards for a stated purpose.

3. Design statistically valid experiments and distinguish repeatability from reproducibility.

4. Quantify measurement uncertainty and recognize important sources of error.

5. Interpret friction, wear, surface, debris, and lubricant data together rather than in isolation.

6. Apply a structured root-cause methodology to tribological failures.

7. Correlate laboratory results with field behavior and identify scale-up limitations.

8. Evaluate the role of automation, digital monitoring, and data analytics in tribological testing.

The Tribosystem as the Unit of Investigation

A tribological result belongs to a system, not to a material in isolation. The tribosystem includes the two bodies in contact, any third bodies such as wear debris or contaminant particles, the lubricant or surrounding medium, and the operating environment. It also includes contact geometry, motion, load, speed, temperature, time, surface topography, material history, alignment, and thermal boundary conditions.

A useful representation is

\[

Y=f(P, Q, M, L, E, T),

\]

where \(Y\) is an observed response such as friction coefficient or wear rate; \(P\) represents operating parameters; \(Q\) surface and geometric characteristics; \(M\) material properties; \(L\) lubricant condition; \(E\) environment; and \(T\) test duration or history. The expression is not normally a complete physical law, but it reminds the investigator that changing one variable may alter the dominant wear mechanism.

For example, a pin-on-disc test may indicate excellent behavior for a coating under constant normal load and dry sliding. The same coating may fail in a lubricated line contact with oscillatory motion because its adhesion, fatigue resistance, and lubricant interaction differ. A test is valid only relative to the question it is intended to answer.

Important terminology follows from this systems view. A screening test ranks alternatives under controlled conditions. A characterization test measures behavior over a defined range of variables. A qualification test determines whether a product meets a specified requirement. A simulation test seeks to reproduce important features of service. A diagnostic test is designed to discriminate between competing failure mechanisms. These categories may overlap, but confusing them can lead to inappropriate conclusions.

From Service Condition to Test Specification

Test design begins with a written problem statement. It should identify the component, contact pair, expected duty cycle, suspected or required performance metric, acceptance criterion, and consequences of failure. The investigator should then construct a service-to-test translation table containing at least:

• contact type: conformal, non-conformal, rolling, sliding, impact, or mixed;
• motion: unidirectional, reciprocating, oscillatory, intermittent, or reversing;
• normal and tangential loads, including transients;
• speed, stroke, frequency, and duty cycle;
• temperature, humidity, atmosphere, and contamination;
• lubricant type, quantity, supply method, degradation state, and replenishment;
• surface material, hardness, roughness, texture, coating, and manufacturing history;
• expected mechanism: adhesive, abrasive, fatigue, corrosive, erosive, fretting, or tribochemical wear.

The dominant similarity requirements should be selected deliberately. Exact similarity in every variable is rarely feasible. For rolling contacts, contact pressure, slide-to-roll ratio, entrainment speed, lubricant viscosity, and temperature may be more significant than matching specimen dimensions. For abrasive wear, particle size, hardness, angularity, concentration, and transport are often critical. For fretting, displacement amplitude, normal load, frequency, and partial-slip conditions may control the result.

A dimensionless approach can help identify the important physics. The film parameter

\[

\Lambda=\frac{h_0}{\sqrt{R_{q1}^2+R_{q2}^2}}

\]

compares lubricant film thickness \(h_0\) with composite surface roughness. A low value indicates greater asperity interaction, although it does not by itself determine the wear mechanism. Other useful groups include ratios of contact pressure to hardness, inertial force to friction force, and thermal generation to heat dissipation. The purpose is not to create unnecessary mathematical complexity, but to prevent superficial matching of speed or load while ignoring the governing contact state.

Tribometers and Instrumentation

Choosing the Test Configuration

The tribometer should reproduce the essential mechanics of the application while permitting reliable measurement and specimen preparation. Common configurations include pin-on-disc and ball-on-disc systems for comparative sliding studies; reciprocating tribometers for coatings, seals, and fretting; block-on-ring and ring-on-ring systems for conformal or line contacts; four-ball and twin-disc machines for lubricants and rolling-sliding contacts; thrust washers for bearing materials; abrasive wheels or slurry devices for three-body wear; and component-level rigs for seals, gears, bearings, and brakes.

Pin-on-disc equipment is versatile but often produces a wear track and pressure distribution unlike those in a real component. Four-ball testing offers standardized lubricant screening, but a ranking obtained in a small point contact should not automatically be treated as a prediction of bearing or gear life. Component simulators are more representative but expensive, slower, and often less suitable for early screening. The correct selection is therefore a compromise among representativeness, control, throughput, cost, and measurement quality.

Measurement Channels

A sound tribological instrument measures more than force. Typical channels include normal load, tangential force, displacement, speed, temperature, vibration, acoustic emission, lubricant pressure, flow rate, humidity, and electrical contact resistance. Optical or capacitive displacement sensors may measure wear depth; torque transducers may measure friction in rotating systems; thermocouples or infrared cameras may capture thermal response; accelerometers and acoustic-emission sensors may detect transitions such as coating delamination or scuffing.

The signal chain consists of the physical phenomenon, sensor, conditioning electronics, analogue-to-digital conversion, software, and data-storage system. Each stage can introduce offset, gain error, noise, drift, aliasing, or saturation. Sampling frequency must be sufficient for the expected dynamics. A low-frequency acquisition system can conceal transient stick-slip or impact events even if the average friction value appears plausible.

Calibration should be traceable to appropriate references wherever practical. Load cells require known-force checks; friction channels require known torque or force; displacement sensors require dimensional standards; thermocouples require temperature references. Calibration before a test is not enough when drift, temperature changes, or contamination can occur during testing. Zero checks, shunt checks, reference specimens, and post-test verification provide evidence that the recorded signal remained credible.

Standards and Their Proper Use

Standards improve comparability by specifying specimen geometry, preparation, operating conditions, reporting, and sometimes calculation methods. ASTM, ISO, DIN, and other national or sector-specific standards are commonly used in tribological work. A standard should be selected according to the purpose of the test, not merely because it is familiar. The investigator must verify the current edition, scope, deviations, and required reporting details.

Standardization does not eliminate engineering judgment. A standardized test may be excellent for ranking materials under a defined condition but weak as a service-life prediction. Any deviation—such as a different surface finish, lubricant volume, counterface, or cleaning procedure—should be recorded because it may affect the mechanism. The test report should state whether the method was used for compliance, screening, research, or simulation.

Experimental Design, Repeatability and Uncertainty

Replication and Randomization

Tribological responses commonly exhibit substantial scatter because surfaces are heterogeneous, debris evolves during testing, and small changes in alignment or preparation can alter the contact. A single test result is therefore weak evidence. Replication estimates random variation, while repeated runs performed on different days, by different operators, or on different machines reveal broader reproducibility.

The number of specimens should be justified from the expected variability, desired confidence, and practically important effect size. If two materials differ by less than the normal scatter of the method, a larger sample or improved control is required before claiming superiority. Randomizing the order of tests reduces confounding with machine warm-up, operator learning, environmental drift, or specimen batch effects. Blocking can be used when tests must be distributed across days, machines, or material lots.

Factorial designs are efficient when several variables may interact. For example, load, speed, and lubricant condition can be varied in a designed matrix rather than one factor at a time. Analysis of variance can identify significant main effects and interactions, while regression or response-surface methods can support optimization within the tested domain. Replication at centre points is useful for estimating pure error and detecting curvature.

Repeatability and Reproducibility

Repeatability describes variation when the same method, operator, apparatus, laboratory, and short-term conditions are used. Reproducibility concerns variation across changed conditions such as laboratories, machines, operators, or time periods. A method can have excellent repeatability but poor reproducibility if small equipment or preparation differences strongly influence the result.

Reporting should include the number of runs, mean, standard deviation, confidence interval, and relevant raw or processed time histories. Friction averages should not conceal run-in behavior, sudden transitions, or unstable periods. Wear rate should be calculated from a clearly defined interval and geometry. If a wear volume is obtained from a profilometer, the treatment of waviness, track edges, debris, and baseline correction must be stated.

Measurement Uncertainty

Measurement uncertainty is not the same as experimental scatter, although both affect confidence in a result. A simplified combined standard uncertainty can be represented as

\[

u_c(y)=\sqrt{\sum_i\left(\frac{\partial y}{\partial x_i}u(x_i)\right)^2},

\]

where \(x_i\) are input quantities and \(u(x_i)\) their standard uncertainties. In practice, contributions may arise from force calibration, displacement resolution, dimensional measurement, temperature, alignment, repeatability, and data processing. Expanded uncertainty is often reported as \(U=k u_c\), with the coverage factor selected according to the intended confidence convention.

Uncertainty budgets should distinguish systematic bias from random variation. Increasing the number of repeats reduces the uncertainty of the mean but does not correct a miscalibrated load cell or an incorrect wear-track model. Sensitivity analysis identifies which input deserves better control. For example, if wear volume is highly sensitive to track depth but not to specimen diameter, profilometry may deserve greater attention than dimensional measurement of the specimen.

Surface, Debris and Lubricant Analysis

Friction and wear data become much more informative when combined with post-test evidence. Surface roughness parameters such as \(R_a\), \(R_q\), skewness, kurtosis, and directional texture can describe preparation, but no single parameter fully represents a tribological surface. Three-dimensional topography, bearing-area curves, waviness, and functional parameters may be needed for seals, bearings, or patterned surfaces.

Optical microscopy provides rapid documentation of tracks, transfer films, polishing, scratches, pits, and delamination. Scanning electron microscopy reveals morphology at higher resolution, while energy-dispersive spectroscopy can identify elemental transfer or contamination. X-ray photoelectron spectroscopy, Raman spectroscopy, and infrared methods can examine thin reaction films and chemical changes. Microhardness, nanoindentation, focused-ion-beam sectioning, and cross-sectional microscopy are useful for coatings and subsurface deformation.

Wear debris is a record of the mechanism. Plate-like particles may indicate severe sliding or delamination; angular particles may suggest abrasion; fine oxide-rich debris may indicate oxidative wear; and fatigue-generated particles may increase as cracks propagate beneath the surface. Interpretation must remain cautious because particle shape changes during transport, collection, and filtration.

Lubricant analysis should address both the intended formulation and its service condition. Relevant measurements may include viscosity, viscosity index, acid or base number, water content, particle count, ferromagnetic debris, elemental composition, oxidation products, additive depletion, and insoluble material. A lubricant can be chemically suitable but mechanically ineffective if its viscosity is too low at operating temperature or if contamination prevents an adequate film. Conversely, a high wear-metal concentration may be a consequence rather than the original cause of failure.

Failure Analysis as a Root-Cause Investigation

A tribological failure investigation begins with evidence preservation. The failed component should be photographed in its installed and disassembled states, labelled, protected from unnecessary cleaning, and accompanied by a chain-of-custody record where legal, warranty, or safety issues exist. The investigator should obtain operating history, maintenance records, lubricant specifications, filter data, load cycles, temperatures, alignment records, manufacturing deviations, and any recent changes in duty or supplier.

A useful investigation sequence is:

1. Define the failure: what function was lost, when, and under what conditions?

2. Separate symptoms from mechanisms. For example, high vibration is a symptom; spalling, looseness, or misalignment may be mechanisms.

3. Establish the damage location, initiation site, progression pattern, and affected population.

4. Compare failed parts with unaffected parts, controls, drawings, and specifications.

5. Form competing hypotheses rather than adopting the first plausible explanation.

6. Test hypotheses using microscopy, dimensional checks, materials analysis, lubricant analysis, and controlled reproduction tests.

7. Identify root, contributing, and latent organizational causes.

8. Verify corrective action through a representative test or monitored field return.

The “five whys” method can expose an immediate chain of causes, but it should not be used mechanically. A fault-tree analysis is useful when several conditions must combine, while a fishbone diagram can organize potential causes under material, machine, method, measurement, environment, and human factors. Failure Mode and Effects Analysis can prioritize risks before failure occurs; it is preventive rather than a substitute for physical evidence after failure.

Mechanism identification requires correlation among morphology, location, operating history, and material response. Adhesive wear often involves transfer and junction formation; abrasion produces directional grooves or cutting; rolling-contact fatigue produces subsurface cracks, pits, or spalls; fretting produces localized debris and oxidized damage under small-amplitude motion; tribocorrosion combines mechanical removal with electrochemical attack. Similar appearances can arise from different causes, so diagnosis should not rest on visual similarity alone.

Correlating Laboratory Tests with Field Performance

A laboratory result becomes valuable for engineering only when its relevance is demonstrated. Correlation may be direct, as when a component simulator reproduces the field mechanism and predicts a measured performance limit. It may also be comparative, when a laboratory ranking consistently distinguishes field-proven alternatives. The correlation statement should identify the variables that were matched, those that were not, and the evidence supporting extrapolation.

Scale effects are important. Contact size changes pressure distribution, thermal dissipation, debris retention, lubricant replenishment, and statistical probability of defects. Laboratory durations may be too short to reproduce incubation and fatigue processes. Accelerated tests may alter the mechanism rather than merely increase its rate. Increasing load until failure, for example, can transform mild oxidative wear into plastic deformation or seizure.

A defensible qualification program often uses stages: screening to eliminate poor candidates, mechanism-focused characterization, component-level simulation, and limited field validation. Field data should be fed back into the laboratory design. If a test predicts friction but not coating delamination, its response metrics are incomplete. Bayesian updating, reliability models, and survival analysis can help combine laboratory evidence with field returns, provided assumptions and censoring are handled correctly.

Applied Case: Premature Spalling in an Industrial Gearbox

Background

This illustrative case concerns a helical gearbox used in a continuous conveying system. The gearbox had operated for approximately one-third of its expected inspection interval when vibration increased and metallic debris was detected in the oil filter. A bearing on the high-speed shaft showed localized spalling on the inner race.

Situation and Initial Hypotheses

Possible causes included inadequate lubricant viscosity, contamination, excessive shaft misalignment, incorrect bearing clearance, overload, material or heat-treatment defects, and mounting damage. Replacing the bearing without distinguishing among these causes risked recurrence.

Analysis

The bearing was photographed before cleaning. Spalling was concentrated in a zone consistent with a loaded contact rather than uniformly distributed around the raceway. Surface examination revealed subsurface-origin fatigue features and secondary polishing of loose fragments. Dimensional inspection found a contact pattern indicating shaft misalignment. Alignment records showed that a coupling had been replaced shortly before the failure, but the alignment verification had not been documented.

The oil had acceptable nominal viscosity, but particle analysis showed a high concentration of hard contaminant particles. Filter debris included angular mineral particles and ferrous fragments. The bearing housing breather was damaged, allowing ingress from a dusty environment. Hardness and metallographic examination did not indicate a manufacturing defect. A laboratory rolling-sliding test using clean oil produced no abnormal damage, whereas tests with representative contaminant particles and the measured misalignment condition produced denting followed by localized fatigue damage.

Approach and Corrective Action

The investigation classified misalignment as a contributing mechanical cause and contamination as a major initiating and accelerating cause. Corrective action included replacement and precision alignment of the coupling, repair of the breather, improved filtration, cleaning of the housing, verification of bearing clearance, and installation of a routine particle-count and ferrous-debris monitoring program. The replacement bearing was not accepted solely on the basis of a short bench test; the system was monitored through a defined observation period under controlled load.

Outcome and Learning Points

Subsequent monitoring showed a declining debris trend and stable vibration. The principal lesson is that the damaged bearing was the location of failure, not necessarily the root cause. A nominal lubricant grade did not prove adequate lubrication, because contamination and alignment changed the contact conditions. The case also demonstrates the value of combining morphology, lubricant analysis, dimensional inspection, maintenance records, and a targeted reproduction test.

Digital Testing and Emerging Practice

Modern tribometers increasingly combine synchronized force, displacement, temperature, imaging, acoustic emission, and vibration data. Automated specimen handling improves throughput, while machine vision can quantify wear-track area or detect coating failure. Digital twins and physics-informed models can link laboratory measurements with component-level predictions, but their reliability depends on validated input data and transparent assumptions.

Machine-learning methods can classify wear debris, identify transitions in friction signals, or detect anomalies in long-duration tests. They are most useful when trained on physically meaningful and well-labelled data. A classifier that recognizes a signal pattern without distinguishing changes in lubricant, sensor condition, or load may produce misleading alarms. Data governance should include calibration records, version-controlled analysis scripts, metadata, and retention of raw signals.

Emerging test practice also emphasizes operando characterization: observing contact temperature, electrical resistance, optical interference, or chemical-film development while the tribosystem operates. Such measurements can reveal transitions that are invisible in post-test inspection. However, adding sensors may alter stiffness, heat flow, or lubrication, so instrumentation itself must be treated as part of the tribosystem.

Challenges, Professional Responsibilities and Best Practices

Common mistakes include using too few specimens, reporting only a mean friction coefficient, ignoring run-in, changing several variables simultaneously, failing to control specimen preparation, overinterpreting a standard test, and treating correlation as causation. Another serious error is cleaning or disassembling a failed component before documenting deposits, debris location, and damage morphology.

Professional practice requires honesty about uncertainty and limitations. Data should not be selectively removed merely because a result is inconvenient; exclusions require documented technical criteria. Safety-critical conclusions should receive independent review. Laboratories should maintain calibration systems, competence records, standard operating procedures, environmental controls, and traceable specimen identification. Confidential industrial data and intellectual property must be protected, while safety-relevant findings should be communicated to responsible parties without concealment.

Best practice can be summarized as follows:

• define the service question before selecting the test;
• describe the complete tribosystem and control the dominant variables;
• use standards where appropriate and document every deviation;
• randomize, replicate, and report uncertainty;
• preserve raw data and calibration evidence;
• combine friction and wear measurements with surface, debris, and lubricant analysis;
• use competing hypotheses in failure investigations;
• distinguish immediate, contributing, and latent causes;
• validate laboratory conclusions against component or field evidence;
• verify corrective action rather than assuming that replacement has solved the problem.

Chapter Summary

Tribological testing is an exercise in controlled representation of a service tribosystem. The appropriate tribometer is determined by contact mechanics, motion, environment, lubrication, and failure mechanism, not by convenience alone. Reliable results require calibrated instrumentation, suitable sampling, controlled specimen preparation, replication, randomization, and explicit treatment of uncertainty. Repeatability and reproducibility describe different aspects of method reliability and must be evaluated separately.

Friction and wear measurements gain meaning when integrated with topography, microscopy, chemical analysis, debris characterization, and lubricant condition. Failure analysis proceeds from evidence preservation and service reconstruction to competing hypotheses, mechanism identification, targeted testing, and verified corrective action. Laboratory-field correlation must account for scale, duty cycle, thermal effects, debris behavior, and possible changes in mechanism during acceleration. Digital sensing, automated analysis, and machine learning offer important opportunities, but only when supported by calibration, validated models, and disciplined data governance.

Key Terms

Tribosystem: The interacting bodies, lubricant, environment, geometry, operating conditions, and third-body materials governing tribological behavior.

Tribometer: An instrument used to impose controlled contact and motion while measuring tribological responses.

Screening test: A comparative test used to rank or eliminate alternatives under defined conditions.

Qualification test: A test used to determine conformity with specified performance requirements.

Simulation test: A test designed to reproduce important service conditions or mechanisms.

Repeatability: Agreement among results obtained under substantially identical short-term conditions.

Reproducibility: Agreement among results obtained under changed operators, equipment, laboratories, or times.

Measurement uncertainty: A quantified indication of the doubt associated with a measured result.

Run-in: The initial period during which surfaces and friction behavior adjust to one another.

Wear rate: Wear volume or another wear measure expressed relative to distance, time, or load-distance.

Third body: Material present between primary contacting surfaces, including debris or contaminants.

Film parameter: The ratio of lubricant film thickness to composite surface roughness.

Tribochemical film: A protective or reactive surface layer formed through chemical processes during contact.

Fretting: Small-amplitude oscillatory damage occurring at contacting interfaces.

Rolling-contact fatigue: Progressive subsurface or surface damage caused by repeated rolling or rolling-sliding stress.

Root cause: A fundamental condition whose removal prevents recurrence of the failure.

Fault tree: A logical model that relates a defined failure event to combinations of contributing events.

Operando measurement: Measurement performed while the tribological contact is functioning under test conditions.

Failure morphology: The physical form and distribution of damage used to infer a failure mechanism.

Review Questions

Short Answer Questions

1. Why is a tribological result considered a property of a tribosystem rather than a material alone?

2. Distinguish repeatability from reproducibility in tribological testing.

3. State two reasons why calibration before a test may not be sufficient.

4. What information can wear debris provide during failure analysis?

5. Why should a laboratory test report document deviations from a standard method?

Long Answer Questions

1. Explain how service conditions should be translated into a laboratory tribological test specification.

2. Discuss the selection of tribometers for screening, characterization, qualification, and component simulation.

3. Explain how replication, randomization, blocking, and uncertainty analysis improve the validity of an experiment.

4. Describe a structured root-cause methodology for a tribological failure investigation.

5. Discuss the challenges involved in correlating accelerated laboratory tests with field performance.

Analytical / Application Questions

1. A coating shows low friction in a dry ball-on-disc test but delaminates rapidly in a lubricated reciprocating seal. Identify the likely limitations of the original test and propose a revised test plan.

2. A gearbox bearing fails after a lubricant change, but the oil viscosity is within specification. Design an investigation using field records, lubricant analysis, surface examination, and targeted laboratory testing.

3. Two materials have mean wear rates of 1.8 and 2.0 units, but their standard deviations are 0.7 and 0.1 units respectively. Explain why the apparent ranking may be unreliable and propose an appropriate experimental decision process.

22. Multiple Choice Questions with Answers

Question 1

Which statement best defines a tribosystem?

A. The hardness of one material measured before testing

B. The contacting bodies, lubricant, environment, geometry, and operating conditions considered together

C. Only the lubricant and its viscosity-temperature relation

D. The software used to calculate wear rate

Correct Answer: B. The contacting bodies, lubricant, environment, geometry, and operating conditions considered together

Question 2

A pin-on-disc test produces highly repeatable results but different laboratories obtain different rankings. This most directly indicates poor:

A. Repeatability

B. Reproducibility

C. Sensor resolution

D. Specimen hardness

Correct Answer: B. Reproducibility

Question 3

What is the principal reason for randomizing the order of tribological tests?

A. To increase the nominal load

B. To eliminate the need for calibration

C. To reduce confounding from time, warm-up, operator, or environmental effects

D. To ensure every specimen has identical wear

Correct Answer: C. To reduce confounding from time, warm-up, operator, or environmental effects

Question 4

Which practice most directly helps distinguish a calibration bias from random experimental scatter?

A. Increasing only the test duration

B. Using calibration checks and repeated measurements

C. Reporting only the maximum friction value

D. Removing all unusual data points

Correct Answer: B. Using calibration checks and repeated measurements

Question 5

A low film parameter generally indicates:

A. Complete separation of surfaces by a thick lubricant film

B. Greater likelihood of asperity interaction

C. Absence of surface roughness

D. That abrasive wear cannot occur

Correct Answer: B. Greater likelihood of asperity interaction

Question 6

Why may a standardized four-ball lubricant test fail to predict actual gearbox life?

A. Standards never allow numerical measurements

B. The small point-contact configuration may not reproduce gearbox geometry, loads, thermal conditions, and duty cycle

C. Four-ball tests measure only temperature

D. Gearboxes do not use lubricants

Correct Answer: B. The small point-contact configuration may not reproduce gearbox geometry, loads, thermal conditions, and duty cycle

Question 7

In a failure investigation, which finding is most useful for separating a failed component from its root cause?

A. Replacing the component immediately without inspection

B. Combining damage morphology with operating history and independent material and lubricant evidence

C. Assuming the newest maintenance activity caused the failure

D. Inspecting only the fracture surface

Correct Answer: B. Combining damage morphology with operating history and independent material and lubricant evidence

Question 8

A wear test is accelerated by greatly increasing load, but the resulting mechanism changes from mild oxidation to seizure. The main interpretive problem is that:

A. The test has become more representative automatically

B. Acceleration has altered the governing failure mechanism

C. The coefficient of friction must be zero

D. Replication is no longer relevant

Correct Answer: B. Acceleration has altered the governing failure mechanism

Question 9

Which measurement is especially useful for detecting lubricant contamination by ferrous wear debris?

A. Particle or ferrographic analysis

B. Only specimen diameter

C. Ambient sound level alone

D. Coating color measurement

Correct Answer: A. Particle or ferrographic analysis

Question 10

What is the most appropriate role of machine learning in tribological monitoring?

A. Replacing calibration and physical interpretation

B. Classifying patterns or detecting anomalies using validated, well-labelled data

C. Guaranteeing prediction outside the training conditions

D. Eliminating the need to preserve raw data

Correct Answer: B. Classifying patterns or detecting anomalies using validated, well-labelled data